Gradient of decomposition in sugarcane mulches of various thicknesses

Crop residues left on the soil surface as mulch influence many services provided in agrosystems, particularly soil protection, water dynamics and nutrient fluxes. For a given crop, the residue mass influences the mulch thickness, but the effect of thickness on decomposition is not well understood, and decomposition gradients within mulches in the field have not been described. This study aimed to quantify the decomposition of sugarcane straw mulches by varying their mass and thickness and determine the decomposition gradient within the mulches. The experiment was conducted on a first-ratoon sugarcane crop for one year, with 4, 8 and 12 Mg straw dry matter ha(-1). Mulches were displayed in 0.16-m(2) litter boxes and formed by straw layers of equivalent mass, i.e., 4 Mg ha(-1), stacked either as a single layer (Low/Top layer), 2 layers (Low and Top layers) or 3 layers (Low, Middle and Top layers) to reconstitute the 3 mulch quantities. We quantified the carbon (C) and nitrogen (N) contents, the chemical composition and the water content of the remaining mulch particles in each layer 10 times within the 360 days. The mulch degradation rates were proportional to the initial amount of straw, the mulch losses representing 75% of its initial C and 46% of its initial N after one year, regardless of the initial mass. Fertilizer-N addition did not change the k decomposition rate, with k = 0.0044 d(-1). However the decomposition rate differed according to the layer position in the mulch, e.g., in the 12 Mg ha(-1) treatment k = 0.0064 d(-1) 0.0046 d(-1) and 0.0033 d(-1) for the low, middle and top layers, respectively, indicating that the low layers in contact with the soil decomposed faster than the top layers in contact with the atmosphere, while the medium layers demonstrated intermediate behaviour. The top layers evolved similarly, regardless of mulch thickness, while the lower layers in contact with the soil decomposed more quickly as the overlying straw thickness increased. Regardless of the thickness and position of a given layer within mulch, the one-year layer C loss was linearly correlated with the average water content of that layer over the year (y = 52.79 - 0694x, R-2 = 0.991, P < 0.05). Here, we demonstrate that the dynamics of mulch decomposition is primarily driven by the dynamics of water, which is itself driven by mulch characteristics, rain regime and evaporation.